636
SHORT COMMUNICATIONS
at least 72 hours at each data point. Over 4oo hours were spent below the alleged
decomposition temperature of 7oo ~ Finally the sample was exposed to a hydrous
atmosphere by means of steam introduced into the furnace for over I2O hours, the
sample itself being at 58o ~
Fig. 1 does not reveal any tendency for the fikermanite to decompose into wollastonite and monticellite. Fig. 2, which is a plot of the atomic spacing of the 211 plane
at various temperatures, also does not indicate any tendency for fikermanite to break
up (if it did, a significant change in the slope would be expected).
The above data tend to indicate that fikermanite, at least in an anhydrous atmosphere, does not, even over a prolonged period of time, decompose into other compounds, but is apparently stable from its melting point down to room temperature.
That the 5kermanite did not decompose at 58o ~ in the increased humidity environment caused by the introduction of steam is of some interest.
Acknowledgements. I would like to thank Mr. W. R. Lees for his aid in setting up the high
temperature equipment.
Texas Tech University
Lubbock, Texas
E. CHRISTIAANDE WYS
REFERENCES
DE WYS (E. C.) and FOSTER(W. R.), 1955. Presented at the 57th annual meeting, American Ceramic
Soc., Cincinnati, Ohio, April 25.
1956. Journ. Amer. Ceram. Soc. 39, 372.
HARKER (R. I.) and TURTLE(O. F.), 1956. Amer. Journ. Sci. 254, 968.
OSBORN (E. F.) and SCHAIRER(J. F.), I941. Amer. dourn. Sci. 239,915.
[Manuscript received 19 April I971]
9 Copyright the Mineralogical Society.
M I N E R A L O G I C A L M A G A Z I N E , MARCH 1 9 7 2 ,
VOL. 38, PP. 6 3 6 - 8
Pyrrhotine and the origin of terrestrial diamonds
HYPOTHESES on the origin of the diamond usually begin with graphite, iron carbide,
or free carbon. This seems appropriate in considering the origin of meteoritic diamonds. However, terrestrial carbon-bearing abyssal magmas, which have surfaced
as kimberlite diamond pipes, are rare. Only one in twenty kimberlite pipes or fissures
contains diamonds (Shand, 1952). Where diamond does occur, special conditions seem
required.
Quite possibly, a carbon or diamond phase originated from the reduction of carbon
dioxide, a common mantle constituent (Hearn, I968; Kennedy and Nordlie, I968).
The reduction could be centered around a specific reducing agent, where its absence
in abyssal magmas could mean the absence of diamond as well. Pyrrhotine could
fill this role. It is a common syngenetic inclusion in South African diamonds (Sharp,
SHORT C O M M U N I C A T I O N S
637
I966). And troilite, another ferrous sulfide mineral, is well known for its close association with meteoritic graphite and diamonds (Lipschutz and Anders, I96I).
At some depth in the earth, diamond will be the thermodynamically stable form of
carbon. This depth, and the corresponding pressure-temperature conditions, can be
determined from the phase diagram for carbon (Bundy et al., I 9 6 0 and pressuredepth and temperature-depth curves for the earth (Clark and Ringwood, 1964). The
geotherms of Clark and Ringwood place the minimum formation temperature between
lO5~176
and I35o ~ The corresponding pressures are 43 and 5I kb, at depths the order
of I4O and i6o km. These conditions may approximate those of actual diamond
formation since graphite inclusions and black diamonds indicate that many diamonds
are formed close to the graphite-diamond equilibrium line.
In this pressure-temperature interval, pyrrhotine could remain below its melting
point (I I95 ~ at 1 atm). In a solid state reaction, however, bulk diffusion rates would
be appreciable.
A possible moderate-temperature-high-pressure reaction between iron-rich pyrrhotine and carbon dioxide is:
2FeS (c)+CO2 (g) ---- 2FeO (soln)+S2 ( g ) §
(diamond).
This open system reaction would be diffusion-c0ntrolled by the carbon dioxide. The
diamond phase could form directly, by-passing a graphite or carbide phase. From
the composition of kimberlite (Kennedy and Nordlie, I968) and the phase diagram
for the system: MgO-FeO-SiO2 (Nafziger and Muan, I967), a separate wtistite or
magnesiowtistite phase is unlikely. The FeO (ideal) would either combine with excess
silica or dissolve in the kimberlitic silicate.
The standard Gibbs energy change for this reaction is 44"64-o'6 kcal at Io5o ~
and 47"2•
kcal at I35o ~ (from the data of Robie and Waldbaum, 1968).
The Gibbs energy change, a measure of reaction feasibility, will be determined by
the standard Gibbs energy change and the actual activity or fugacity of each substance.
With an activity coefficient, 7, of o-i, the activity of FeO at a mole fraction of o'o5
(Kennedy and Nordlie, I968) would be o'o05. This activity coefficient is quite reasonable for a silicate solution since a partial molar heat of solution AH, as small as
--6 kcal per mole at IO5o ~ could give this value (4"58 T l o g 7 ~ M4). The contribution this activity makes to the Gibbs energy change is --27. 7 kcal at lo5o ~ and
--34"o kcal at I35o ~
The net effect of the prevailing pressure on the solid state activity will be determined
by the reaction molal volume change. This could be as large as --9"o cm 3 (Robie and
Waldbaum, I968), neglecting mineral compressibilifies and thermal expansions. The
contribution to the Gibbs energy change would be --9 kcal at IO5O ~ and - - I I kcal
at I35o ~
Then, assuming CO2 and $2 have the same fugacity coefficient, a $2 mole fraction
the order of o.I would be adequate for the reaction to proceed in the temperature
interval Io5o to I35O ~
In summary, a new hypothesis is proposed for the formation of terrestrial diamonds
with carbon dioxide as the source of carbon, and pyrrhotine as the reducing agent and
Tt
638
SHORT COMMUNICATIONS
r e a c t i o n catalyst. Kinetically, the direct conversion o f c a r b o n dioxide to d i a m o n d
overcomes certain experimental p r o b l e m s ( K e n n e d y a n d N o r d l i e , I968).
Aerospace Corporation,
PAUL C. MARX
El Segundo, California
REFERENCES
BUNDY (F. P.), BOVENKERK(H. P.), STRONG(H. M.), and WENTORE(R. H., Jr.), I96I. Journ. Chem.
Phys. 35, 383-9I.
CLARK (S. P.) and RINGWOOD(A. E.), ~964. Rev. Geophys. 2, 35-88.
HEARN (B. C., Jr.), 1968. Science, 159, 622- 5.
KENNEDY (G. C.) and NORDLIE(B. E.), 1968. Econ. Geol. 63, 495-5o3.
LIPSCHUTZ (M. E.) and ANDERS(E.), I961. Geochimica Acta, 24, 83-1o5.
NAFZIGER(R. H.) and MUAN (A.), 1967. Amer. Min. 52, 1364-85.
ROB~E (R. A.) and WALDBAUM(D. R.), I968. Thermodynamic properties of minerals and related
substances at 298"I5 ~ and one atmosphere and at higher temperatures. U.S. Geol. Surv. Bull.
1259.
SHAND (S. J.), I952. Rocks for chemists. New York (Pitman).
SHARP (W. E.), 1966. Nature, 211, 4o2- 3.
[Manuscript received T July I97I]
9 Copyright the Mineralogical Society.